This study aims to examine the nutritional, metabolic and growth effects of antiretroviral therapy (ART) on children in South Africa, the country with the largest number of children requiring ART in the world. Currently little is known on the effects of ART in children, and what is known comes from industrialized countries, where the malnutrition and infectious co-morbidities so prevalent in African children are often not present. The findings from this study will be important in informing both the clinical management and public health policy of the worldwide initiative to provide ART to the world's HIV-infected children. Specific aims of this study are to: 1) describe growth and body composition changes in the 2 years following initiation of ART;2) describe the incidence of metabolic derangements and lipodystrophy in these children and temporal relationship with commencement of ART;3) describe the relationship between pretreatment nutritional status and drug tolerability/ metabolic abnormalities in the 2 years after initiation of ART;4) determine if body composition changes are related to changes in viral load or CD4 percentage;5) determine predictors of ART failure. We will recruit 150 treatment naive symptomatic prepubertal HIV-infected children who are commencing triple-drug antiretroviral therapy. Study participants will be recruited during the first 2 years of the study, and will be recruited from 2 sites: the HIV Family Clinic in King Edward VIII Hospital in Durban;and at the Africa Centre/Hlabisa Health sub-District ART program in rural KwaZulu Natal. ART will be provided according to ART guidelines of the KwaZulu Natal Provincial Department of Health (KZN DOH). These guidelines recommend stavudine, lamivudine and lopinavir/ritonavir for children less than 3 years, or stavudine, lamivudine and efavirenz if older than 3. Study participants will be followed monthly initially and then 3 monthly. We will measure attained growth (height and weight), body composition (skinfold thickness and bioimpedance assessment), lipodystrophy (clinical features and skinfold ratio), markers of dyslipidemia and glucose homeostasis, evidence of drug adverse effects and evidence of treatment failure (based on changes in CD4 cell percentage and HIV viral load) at each visit throughout the study. We will use the following statistics to describe growth patterns and frequency of metabolic abnormalities following initiation of ART: means/medians and 95% confidence intervals, bivariate and multivariate logistics regression models to determine the relationship of pretreatment nutritional status, drug tolerability and metabolic abnormalities, and predictors of treatment failure. All data will be entered into Epi Info and will be analyzed using standard PC-based data analysis packages, including SAS and SPSS.